JPH0481974A - Design device - Google Patents

Abstract

PURPOSE:To obtain an exact design by inferring and outputting the design by inputting image of a design object which a user, etc., have vaguely. CONSTITUTION:The design device is constituted of a computer1, a keyboard 3 and a CRT 5, and the computer 1 is provided with a memory M being a knowledge data base CL1 and a CPU 7 being a design inference means CL3. In such a state, when an image of a design object is inputted by an image input means CL2, a design is inferred by the design inference means CL3, based on the inputted image and the knowledge data base CL1. This inferred design is outputted by a design output means CL4. In such a way, a user's image can be inputted exactly.

Description

[Detailed Description of the Invention] [Objective of the Invention] (Industrial Application Field) This invention is applicable to displaying the design of a vehicle on an output means such as a CRT while confirming the image that a person has of the vehicle. Regarding design equipment.

(Prior art) This type of design device includes, for example, March 1988: Published by the Institute of Electronics, Information and Communication Engineers, “Journal of the Institute of Electronics, Information Engineers: Separate Volume: Vol.
1, 71NO, 3pp, 245-247J, as shown in FIG. This is HULIS (
It shows the configuration diagram of the Human, LivLng System), and the exterior of the house is used as an adjective to express the house design elements and feeling for each component such as the entrance, Western-style room, Japanese-style room, kitchen, bath, etc. Evaluation terms such as ``luxury,''``spacious,'' and ``comfortable,'' knowledge that defines the relationships between evaluation terms, and knowledge that defines the relationships between these evaluation terms and design elements are stored as a knowledge database, and human operates an evaluation terminology input means such as a keyboard, and inputs, for example, a limited term for housing design such as ``Japanese-style room,'' which represents a design element that is a component of a house, and also inputs his or her own image of the Japanese-style room. When the evaluation term "spacious" representing the ring is input into the computer, the computer infers and extracts appropriate Japanese-style room design elements from the input evaluation term from the knowledge database and displays the image on the display means. .

By the way, if the technical evaluation terminology of the above-mentioned design device is adapted to the vehicle design, it can be put to practical use as a vehicle design device.

That is, evaluation terms as adjectives expressing the feeling of the vehicle, knowledge that defines the relationship between the evaluation terms, vehicle design elements, and knowledge that defines the relationship between the evaluation terms and the design elements are stored in the memory of the computer.

The evaluation terms are composed of adjective words as shown between the evaluation terms in FIG.

The knowledge that defines the relationship between evaluation terms is the knowledge that was obtained by collecting evaluation terms and analyzing the results of preliminary experiments as shown in FIG. 10. This relationship refers to, for example, dividing evaluation terms into factors based on human intervals from 1 to 10 using quantification theory class I or type II multivariate analysis, and determining how much relationship each evaluation term has with each factor. This is data obtained as factor loadings (values). It can be said that terms with numbers that have similar loadings on each factor are closely related and similar. When this knowledge that defines the relationship between evaluation terms requires a design device, if the input evaluation term is not among the evaluation term group used in the sensory evaluation experiment described later, the most similar term is selected. used for.

As shown in FIG. 11, the vehicle design element group is divided into categories for each item, such as the number of meters in the vehicle interior.

The knowledge that defines the relationship between evaluation terms and vehicle design elements was obtained as a result of the sensory evaluation experiment shown in FIG. 11 above. Specifically, items for evaluation terms and vehicle design elements are selected, vehicle designs corresponding to each item are shown to an unspecified number of people, and each evaluation term, which is a group of adjective words, is received from the vehicle design elements. Collect feelings and calculate partial regression coefficient (
It is analyzed as a correlation coefficient).

Here, the items include the number of meters installed on the instrument panel, the instrument panel and meter cluster, the meter cluster and center console, the thickness of the door, the overhang of the center cluster, and the armrest, as shown in Figure 11. The focus is on each of the following.

The number of meters is classified into five categories. That is, the first category has one large meter, the second category has one large meter and two small meters, the third category has one large meter and three small meters, and the second category has two large meters. The fourth category includes two large meters and one small meter, and the fifth category includes two large meters and two small meters.

Items focusing on the instrument panel and meter cluster are divided into two categories. That is, there are a first category in which the instrument panel and the meter cluster are integrated, and a second category in which the instrument panel and the meter cluster are separated.

Items focusing on the meter cluster and center console are divided into two categories.

That is, a first category includes a meter cluster and a center console, and a second category includes a meter cluster and a center console.

Items that focus on the thickness of the door are classified into two categories. That is, the first category has thick doors, and the second category has thin doors.

Items that focus on the overhang of the center cluster are classified into two categories. That is, the first category has a center cluster overhang, and the second category has no center cluster overhang.

Below, items focusing on armrests and other items are classified into categories.

In this state, if you operate an evaluation term input means such as a keyboard and input an evaluation term such as "luxury," the computer will calculate the partial regression coefficient for each item based on the knowledge that defines the relationship between the evaluation term and the vehicle design element. The category (indicated by the east mark in Fig. 11) for which the
)(b)). For example, if two evaluation terms such as "luxurious" and "spacious" are input, the computer will sum up the partial regression coefficients for each category for each item of the two terms, and then calculate the partial regression coefficient for each item. The category with the highest total value is extracted and displayed as an image of the vehicle interior design.

(Problems to be Solved by the Invention) By the way, in such a design device, the input is an adjective word as an evaluation term, and the evaluation term is directly input. However, when a user actually uses this device, the image that the user has is not limited to adjective words, so it is difficult to accurately express and input the vague image that the user has of the vehicle interior. Ta.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a design device that can match the image that a user has with vehicle interior evaluation terms and items, and can more accurately input the user's image.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above object, the present invention, as shown in FIG. The configuration includes means CL2 for inputting an image, means CL3 for inferring a design based on the input image, and means CL4 for outputting the inferred design.

(Operation) According to the above configuration, when an image of a design object is inputted by the image inputting means CL2, a design is inferred by the design inference means CL3 based on the inputted image and the knowledge database CLI. This inferred design is output by the design output means CL4.

(Example) Embodiments of the present invention will be described below based on the drawings.

FIG. 2 shows a schematic configuration diagram according to an embodiment of the present invention, and this design device is composed of a computer 1, a keyboard 3, and a CRT 5.

The computer 1 includes a memory M as a knowledge database CL1 and a CPU 7 as a design inference means CL3.

Here, memory M has evaluation term 9. Knowledge that defines the relationship between evaluation terms 11. Vehicle design elements 13, knowledge 15 that defines the relationship between evaluation terms and design elements, and options 17 that express the image of the vehicle interior are stored, respectively.

The evaluation terms 9 are composed of adjective words as shown between the evaluation terms in FIG.

Knowledge 11 that defines the relationship between evaluation terms is knowledge obtained by collecting evaluation terms 9 and analyzing the results of a preliminary experiment, as shown in FIG.

As shown in FIG. 11, the vehicle design element group 13 is divided into categories for each item, such as the number of meters in the vehicle interior.

Knowledge that defines the relationship between evaluation terms and vehicle design elements! a
No. 15 was obtained as a result of the sensory evaluation experiment as shown in FIG.

Option 17 for expressing the image of the vehicle interior expresses the image using a plurality of different pictures, photographs, colors, sounds, scents, etc.

The keyboard 3 constitutes an image input means cL1 for selecting and inputting an image option 17, and constitutes a term input means for inputting an evaluation term 9.

The CPU 7 infers a corresponding evaluation term from the knowledge database based on the input image choices, and
The design is inferred from the knowledge database based on the input evaluation terms.

The CRT 5 constitutes a display means for displaying image options, and also constitutes a design output means CL3 for outputting the inferred design.

According to the above configuration, 9. Pre-selected evaluation terms; 11. Knowledge that defines the relationship between evaluation terms obtained in a preliminary experiment; Vehicle design elements 13. Knowledge that defines the relationship between evaluation terms obtained through sensory experiments and vehicle design elements 215. Choices 17 and the like expressing the image of the vehicle interior are stored in the memory M of the computer 1. In this state, by operating the keyboard 3, a selection 17 representing an image of the vehicle interior is displayed on the CRT 5. When an option matching the image is selected from the displayed screen using the keyboard 3, the CPU 7 infers an evaluation term corresponding to the option from the knowledge database and displays it on the CRT 5. When you enter as many evaluation terms 9 as you like using the keyboard 3 while looking at the displayed screen, each vehicle design element 13 is inferred from the knowledge database by the CPU 7 of the computer 1.
It is displayed on the CRT 5 as an extracted design drawing.

The following explanation will be given based on the flowchart of FIG.
5 is displayed. For example, in step S1, "Car A", "
A case is shown in which photos of vehicle interiors of "Car B" and "Car C" are displayed.

Here, an explanation will be given assuming that the user has an image of selecting a sports car-type vehicle interior with an active nature. In this case, the user selects "Car B", which has a sports car type and is active, from among the photos displayed on the CRT 5 and inputs it using the keyboard 3. Then, by the action of the CPU 7 of the computer 1, "for runners", "exciting", "active", "sporty", "wild", etc., which contribute highly to the activity factor, are selected from the knowledge database. The evaluation term 9 is inferred and displayed on the CRT 5.

This inference process is shown in FIGS. 4 and 5.

Figure 4 shows the correspondence between options and evaluation terms based on the results of factor analysis. In other words, based on the arrangement of the factor scores and factor loadings for each factor axis obtained through the factor analysis in Evaluation Terminology 9, select the option with the factor score that most represents each factor axis, and calculate the factor loadings for that factor axis. A term with a high value becomes a corresponding evaluation term. For example, in Figure 4, "Car A" is placed in the vehicle interior representing the factor axis of "calmness", and evaluation terms such as "never boring", "comfortable", "elegant", etc. are associated with it. . In addition, the "B car" is placed in the vehicle interior that represents the factor axis of "activity", and the evaluation terms are "exciting", "active",
"Sporty" etc. are associated with each other. Similar correspondence is made for other factor axes.

FIG. 5 shows the process of inferring evaluation terms corresponding to options. For example, if a user imagines a sports car-type vehicle interior, the user selects photos "Car A" and "Car B."
, if a photograph of the vehicle interior of "Car B" representing the activity factor is selected from "Car C", the CPU 7f) of computer 1 will calculate the activity factor based on the factor loadings in Figure 10. ``For runners'', ``exciting'', ``active'', and ``sporty'' that contribute greatly to
, “wild”, etc. are inferred, and the CRT
Displayed on 5.

Next, in step S2 in FIG. 183, when the user inputs, for example, an adjective word such as "exciting" or "sporty" from the displayed evaluation term group as an evaluation term that matches the image, by operating the keyboard 3, In the next step S3, the CPU 7 of the combiator 1 infers and extracts the category with the maximum total value of partial regression coefficients for each item from the knowledge database, and displays the image on the CRT 5 as the vehicle interior design (step S4). .

Therefore, it is possible to match the vague image that the user has with the evaluation term for the vehicle interior, it is possible to input the image that the user has more accurately, and it is possible to bring the design of the vehicle interior closer to the user's image. can.

6 to 8 show flowcharts according to other embodiments of the present invention. Steps similar to those in FIG. 3 are given the same reference numerals and redundant explanations will be omitted.

In the embodiment shown in FIG. 6, music such as "classical,""jazz," and "popular" are displayed on the CRT 5 as options 17 for expressing images in step S11.

Here, for example, select "Classic" and use Keyboard 3.
When input, the CPU 7 of computer 1 uses the knowledge database to select evaluation terms that match the image of "classic" such as "calm,""elegant,""heavy,""luxury," etc. 9 is inferred and displayed on the CRT5.

The user refers to the evaluation term group displayed on the CRT 5 and inputs terms that match the image using the keyboard.

In the embodiment of FIG. 7, the choices 17 for expressing the image in step S21 include "red", "yellow", "blue", "green",
A color such as "black" is displayed on the CRT 5.

For example, when you select "red" and enter it on the keyboard 3, the CPU 7 of the computer 1 selects "active", "lively", "bright", etc. that match the image of "red" from the knowledge database. An evaluation term 9 such as "lively" is inferred and displayed on the CRT 5.

In the embodiment shown in FIG. 8, step S31 is added to the flowcharts shown in FIGS. 3, 186, and 7. , the user can choose what to express, such as color, blue, scent, etc.

Therefore, the embodiment shown in FIG. 6 can be designed with a sound image, and the embodiment shown in FIG. 7 can be designed with a color image.
In the embodiment shown in FIG. 8, it is also possible to select the type of image to be input.

Furthermore, in each of the above embodiments, option 1 to express the image
Although one image is selected from 7, two or more images may be selected. It can be configured to infer a design directly from the input image.

[Effects of the Invention] As is clear from the above description, according to the configuration of the present invention, a design is inferred and output by inputting a user's vague image of the design target, so that more accurate design can be achieved. You can get the design.

[Brief Description of the Drawings] Fig. 1 is a block diagram of the present invention, Fig. 2 is a block diagram of an embodiment of the invention, Fig. 3 is a flowchart of an embodiment, and Fig. 4 is a factor analysis result. FIG. 5 is an explanatory diagram showing the inference process of evaluation terms corresponding to the options. FIGS. 6 to 8 are flowcharts according to other embodiments. Figure 10 is a block diagram of the conventional example, Figure 10 is a table of knowledge that defines the relationship between evaluation terms, Figure 11 is a table of sensory evaluation experiment results, and Figures 12 (a) and (b) are of CRT. It is a figure which shows an example of an output. CLl...Knowledge database CL2...Image input means CL3...Design inference means CL4...Design output means

Claims (2)

[Claims] (1) A knowledge database of the relationship between design images and design elements, a means for inputting an image of a design object, a means for inferring a design based on the input image, and a means for outputting the inferred design. A design device equipped with (2) A knowledge database of evaluation terms expressing the design and the relationship between design elements, means for inputting images of design objects, means for inferring evaluation terms from the input images, and outputting the inferred evaluation terms. a means for selecting a specific evaluation term from the output evaluation terms; a means for inferring a design based on the selected evaluation term and the knowledge database; and a means for outputting the inferred design. Featured design equipment.